This file was created by scanning the printed publication.
Errors identified by the software have been corrected;
however, some errors may remain.
Chapter32
Environmental Risk Assesslnent and Deployment Strategies
for Genetically Engineered Insect-Resistant Populus1
Kenneth F. Raffa, Karl W. Kleiner, David D. Ellis, and Brent H. McCown
Introduction
Most studies on genetically engineered plants have concentrated on efficacy; few have focused on environmental
safety {Seidler and Levin 1994). This emphasis reflects the
rapid increase in our technological capabilities over the last
15 years and reflects an uncertainty over how best to scientifically generate relative rankings of the likelihood and severity of possible adverse effects. Environmental risk
assessment is not an exact science and can only provide direct comparisons between treatments and checks under the
most controlled, and therefore least realistic, conditions.
Moreover, risk assessment inevitably raises questions that
are at least partially subjective, value-laden, and contextual.
This chapter attempts to ·identify the more significant
issues of environmental risk, weigh their relative impacts,
and suggest possible strategies for reducing adverse effects. We limit our discussion to potential environmental
effects. Impacts on social and economic systems, while
important, are beyond our expertise.
Approaches to Environmental
Risk Assessment
An appropriate definition of environmental risk is debatable (Morgan 1993; Wilson and Crouch 1987). In this
1
Klopfenstein, N.B.; Chun, Y. W.; Kim, M.-S.; Ahuja, M.A., eds.
Dillon, M.C.; Carman. R.C.; Eskew. L.G., tech. eds. 1997.
Micropropagation. genetic engineering. and molecular biology
of Populus. Gen. Tech. Rep. RM-GTR-297. Fort Collins, CO:
U.S. Department of Agriculture, Forest Service, Rocky Mountain
Research Station. 326 p.
chapter, environmental risk means: "the likelihood that
release of a novel material will cause adverse effects such
as mortality or reduction in populations of nontarget organisms due to acute, chronic, or reproductive effects, or
disruption of community or ecosystem function" {Urban
and Cook 1986). Predicting consequences becomes increasingly complex as the scale expands from individual- to
community-level interactions. For example, negative effects
on individuals do not necessarily translate into reduced
populations. Natural and managed systems provide many
instances of compensatory feedback where removing
substantial numbers of individuals does not affect population density. Conversely, the prospect of indirect effects
from community-level interactions, even when no individual effects appear important, is a serious concern. Basic ecological studies provide a wealth of examples.
Indirect interactions, across multiple trophic levels, incorporating biotic and abiotic environmental factors, and
mediated by a broad range of symbionts, competitors, and
alternate hosts, can exert enormous influences and generate unpredicted outcomes {Angle 1994; Bergelson 1994;
Ehler 1990; Holt 1977; Price et al. 1980; Simberloff 1985).
Similarly, the history of applied ecologies, such as agriculture and forestry, shows that indirect interactions often yield the least predictable yet most damaging
consequences.
The issue of ecological complexity is a paradox to risk
assessment. Controlled evaluations of acute effects on isolated individuals generate the least variable and seemingly
most "reliable" results. Studies that attempt to unravel
more diffuse and incipient effects are ultimately more important, yet unfortunately they are less likely to provide
definitive answers {Angle 1994). Ironically, current approaches to training, funding, and productivity are
strongly biased toward the former approach.
Comprehensive risk assessment must also weigh anticipated benefits against risk. Genetic engineering of Populus
offers several potential benefits, such as reduced pesticidal
inputs, improved carbon sequestration, alleviation of pressures to exploit unmanaged systems, and improved
sources of alternatives to fossil fuels {Kleiner et al. 1995;
249
Section V Biotechnological Applications
McCown et al. 1991; Raffa 1989; Raffa et al. 1993; Robison
et al. 1994). These are potentially enormous benefits; however, they are beyond the scope of this paper.
Potential risks can be categorized based on their spatial
and temporal scales. For example, an effect could be limited to the treated site, or it could impact neighboring ecosystems. Effects can be short-term, such as the release of a
toxic gene product into the environment, or self-replicating, such as the escape of viable germplasm. Such distinctions can be somewhat blurred and need to be assessed as
part of the complete risk evaluation process. Still, there is
general agreement that the most serious concerns arise
when genetically engineered organisms could cause selfperpetuating injury to commercial or natural ecosystems
beyond the immediate area of release.
Criteria for Risk Assessment
Different individuals, agencies, and organizations have
advocated different criteria, burdens of proof, and levels
of evidence governing the planned release of genetically
engineered organisms. Probably the most helpful guidance
is provided by a National Academy of Sciences committee headed by Arthur Kelman (NAS 1987). In the opinion
of NAS, what matters is the product not the process. According to this perspective, introducing genetically engineered organisms "poses no risks different from the
introduction of unmodified organisms and organisms
modified by other methods." Therefore, "assessment of risk
should be based on the organism, not the method of engineering." Subsequent authors have delineated some important differences between g~netic engineering and plant
breeding, and hence the need for limits in applying this
equivalency (Dale and Irwin 1995; Giampietro 1994; Regal 1994). However, this starting point has proven useful
and has withstood the test of time. Similar conclusions are
stated by Tiedje et al. (1989) in an Ecological Society of
America report: "transgenic organisms should be evaluated and regulated according to their biological properties (phenotypes), rather than according to the genetic
techniques used to produce them."
Emphasizing phenotypes over the methods by which
they arise has proven useful because it rebuts scientifically
unfounded criticisms and focuses on interactions between
gene products and their environment. Rather than dismissing environmental concerns, this approach highlights the
importance and complexity of predicting responses to gene
products at the community level, and the need for ecological expertise in devising scientifically based policies.
If the product not the process is critical, then expertise in
the methods of genetic engineering is not directly relevant
to predicting how novel organisms will interact with eco-
250
systems. Molecular expertise is invaluable, however, in
protecting against unintended changes in the genome, incorporating methods of sterility, and controlling and evaluating patterns of expression. The criteria for estimating and
the approaches to alleviating environmental concerns require interdisciplinary efforts (Raffa 1989).
Raising every imaginable hazard that could arise from
genetically engineered organisms is neither difficult nor
helpful. This approach could hinder the enormous value
of biotechnology and dilute needed emphasis on legitimate concerns. At the other extreme, the view is sometimes expressed (or implied) that all concerns arise merely
from a lack of scientific understanding or breadth. This
view seriously underestimates the complexity of scaling
from molecular- through ecosystem- level processes. Failure to consider such complexity invariably detracts from
the long-term sustainability of new technologies; a costly
lesson already appreciated by agrichemical companies. The
issue needs to be one of reasonable probability (de Zoeten
1991; Frederick and Egan 1994; Hubbes 1993; NAS 1989;
Raffa 1989; Strauss et al. 1991; Tiedje et al. 1989). For example, Tolin and Vidaver (1989) propose that "restrictions
should be based on demonstrated, not conjectural risks."
However, we would substitute "realistic" for "demonstrated" to promote a more proactive approach· to risk
management. In our view, the likelihood of risk may be
realistic if 2 conditions are met: 1) a clear mechanism, based
on known biological processes and verified assumptions,
can be delineated; and 2) there is relevant precedent.
Few specific risks meet the above criteria. Those that do
can be classified into 3 general categories: 1) escaped plants
or genes, 2) evolution and consequences of resistant pest
biotypes, and 3) alteration of multi-trophic processes. We
first describe how Populus systems relate to these questions and then address each risk. Biotype evolution will
be developed as a more detailed case study, as this is our
primary area of interest. We conclude with an overall synthesis of environmental risk assessment in Populus.
For each concern, the potential risk can be addressed
by asking 3 questions: 1) "Is there assurance that the proposed event (i.e., gene escape, biotype evolution, altered
multi-trophic process) will not occur?"; 2) "Is there assurance that the effects will be harmless if this event does
occur?"; and 3) "Are there ways for reducing the likelihood and impact of harmful effects?" These questions
place the burden of proof on the novel gene product to be
consistent with ~ow other novel products such as biological control agents, introduced plant materials, and pesticides are evaluated (Caltagirone and Huffaker 1980;
Charudattan and Browning 1992; FIFRA 1978; Fuester
1993; Harris 1985; Hinkle 1993; Hutton 1992; Upholt 1985;
USDA FS 1991; White et al. 1992). During actual experimentation, however, the null hypothesis is one of no effect. Also note that questions 1) and 2) are evaluated in
the absence of any ameliorating steps, but the availability
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
and implementation of such tactics (question 3) could
greatly reduce resulting concerns.
Populus Growing Systems
The safety of releasing a genetically engineered organism is determined by the gene product and the environment into which it will be introduced (Abbott 1994; Falk
and Bruening 1994;}epsonet al. 1994; Pimentel et al. 1989).
In general, risks are lower for intensively cultured short
rotation tree crops than for large forested expanses of longlived species (Raffa 1989). Populus occupies a relatively
broad range of growing conditions along this continuum.
For example, trembling aspen, Populus tremuloides, is one
of the most widely distributed naturally occurring species in North America. It is a valued forest tree with anumber of uses such as soil quality improvement, watershed
maintenance, C02 sequestration, and wildlife habitat.
When used for timber, Populus is harvested from self-regenerating forests and grown in commercial stands.
Populus is also a major component of rapid rotation systems such as biofuel plantations. These intensively managed systems, more closely resembling agricultural than
forest production, have short growing intervals, are based
on carefully derived clonal material, and are subjected to
intensive cultural and chemical inputs.
Populus has also become the focus of intense basic research by molecular biologists, plant physiologists, and
ecologists. Populus is the preeminent tree model for tissue
culture, molecular mapping, and transgenic technology.
Concurrently, Populus has become a key model for basic
ecophysiological and plant-herbivore interaction studies.
Thus, Populus provides an ideal system for evaluating the
role of plant community structure in the efficacy and environmental safety of various deployment strategies and
for integrative studies from the molecular- through ecosystem- level scales.
Movement of Transgenes Into
Native Populations
The most direct form of proposed environmental harm
is that genes encoding novel traits might become established in feral populations and subsequently exert a weedy
effect. A variety of mechanisms for gene escape have been
proposed such as hybridization into gene pools of wild
relatives, crop abandonment, movement of cuttings by
animals or water, etc. Escape of transgenes into the envi-
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
ronment could occur on several levels, and the risk associated with each level of escape varies. A commonly proposed level of escaped transgenes is via pollen or seed.
Another level is the physical movement of plant parts into
the surrounding environment. Because virtually all the
poplars used in short rotation plantations are clones multiplied by vegetative cuttings, small branches can be moved
off site by animals or overland water flow. The root sprouting habit of several poplars poses a similar concern.
Precedents by which to evaluate possible consequences
of escaped material are offered by the literature on accidental or injudicious introduction of exotic species. Following the N AS (1987) rationale that introducing
genetically engineered organisms should not be treated
differently from other unmodified organisms, the literature on exotic introductions provides useful precedents to
evaluate the likelihoods and consequences of escapees
(Williamson 1994). A large and well-documented data base
details numerous instances in which accidentally or deliberately introduced microorganisms, arthropods, nematodes, plants, and vertebrates became established in
non-native ecosystems, and subsequently exerted severe
economic and environmental consequences (e.g., Liebhold
et al. 1995; Lodge 1993; Mooney and Drake 1986; US Congress OTA 1993 ).
A number of molecular biologists, agronomists, ecologists, and plant protection specialists have 'cautioned that
some escape is likely. For example, Strauss et al. (1995)
stated unambiguously, uGene flow within and among tree
populations is usually extensive, which makes the probability of transgene escape from plantations high."
Timmons et al. (1995, 1996) expressed a similar conclusion
for Brassica. Likewise, the ecologists Kareiva et al. (1994)
concluded that "the escape of transgenic pollen is inevitable." Williamson's (1994) analysis of historical records
of deliberately introduced organisms, concludes that
nearly all escape, and of these 10 percent become established.
Our ability to address whether there is sufficient assurance that escaped genes would be harmless is considered
case by case (Dale and Irwin 1995). Various authors, including molecular biologists (Strauss et al. 1995), ecologists (Kareiva et al. 1994; Regal 1994; Seidler and Levin
1994; Tiedje et al. 1989), and crop protection specialists
(Dale 1994; Williamson 1994), have identified possible
adverse effects of escaped transgenes. Some examples include creation of new (or enhanced) pests, harm to nontarget species, and disruptions to biotic communities,
natural food webs, and ecosystem processes.
In each of these cases, there are well established mechanisms by which such adverse consequences might arise,
and substantial literature providing precedents from analogous introductions. Examples of possible mechanisms include: 1) enhanced competitiveness of a genetically
engineered organism (due to pest resistance or physiologi-
251
Section V Biotechnological Applications
cal environmental tolerance of stress) that displaces existing or subsequent beneficial organisms (Ellis et al. 1984;
Moamad et al. 1984); 2) reductions in seed dispersal, pollination, or biodiversity by insecticidal transgene products
(Simmonds 1976; McGranahan et al.·1988); or 3) acquisition of traits that enhance competitive status by existing
weed species (Windle and Franz 1979). Again, the historical record with traditional introductions is of some value.
Williamson (1994) reports that 10 percent of the 10 percent
of escaped species that establish become problematic. It
must be emphasized,. however, that such figures do not
reflect refined deployment strategies (using specific information about target species, transgenes, rotation cropping
system, and location) that could accompany planned releases of transgenic poplars. And, although the historical
record of planned releases of genetically engineered organisms is still relatively small, to date there have been no
known adverse effects.
Despite the potential for adverse effects, a number of
attributes of the transgene, the parent organisms, phenotypic expression, and target pest-environment system
could reduce risk (Tiedje et al. 1989). For example, risk
analysis must consider whether pollen from transgenic
hybrids is compatible with surrounding populations and
also whether the timing of pollen release occurs when stigmatic surfaces in surrounding populations are receptive.
Any impact of escaped genes will likely vary with the novel
gene product as well. For example, risks associated with
the escape of a Bacillus thuringiensis (Bt) endotoxin gene
may be different than those for a gene modifying lignin.
Differences in plant species and growing systems are also
pertinent. In agronomic food crops such as soybean, maize,
potato, and tomato, measures such as sterility have not
been a requirement for registration. Conversely, a case-bycase analysis of each transgene-species-planting site, combinations may be needed.
The third question, whether the risk or impact of escape
can be ameliorated, is currently the subject of intense effort. Risk from escape by vegetative material could be reduced by management practices that minimize root
sprouting outside the plantation and the distance plant
material is moved. This can be achieved by planting buffer
strips that are routinely cultivated and/ or planted with
an annual crop so that escapes can be readily identified
and treated with herbicide. One management strategy for
contending with pollen or seed dispersal might be to identify late flowering clones for a breeding population, such
that harvesting occurs before sexual maturity in the
transgenic trees. Such an approach would offer functional,
while not physiological, sterility. Another strategy might
be the use of sterile triploids.
Physiological approaches to reproductive sterility in
genetically engineered trees have recently been reviewed
(Strauss et al. 1995) and are not be treated extensively here.
Basically, these include using floral promoter-cytotoxin' to
252
ablate floral tissues and disrupting expression of essential
floral genes. In the former approach, cytotoxic genes regulated by reproductive-specific promoters kill all cells committed to reproductive development. The latter approach
uses antisense RNA, sense suppression, or promoter-based
suppression to impair the expression of genes required for
fertility. These approaches can be deployed with varying
levels of gender specificity, have relative advantages and
disadvantages (Meilan and Strauss this volume; Strauss
et al. 1995), and have yielded some successes (Mariani et
al. 1990). The current obstacles relate to our lack of basic
information about reproductive gene sequences and expression in clonally propagated species such as Populus.
Very little is known about the long-term stability of
transgene expression in woody perennials. In 1 test examining the expression of a marker gene in field poplars, some
level of seasonal variability in transgene expression was
observed, but in general this variation was predictable, and
relatively continuous expression levels occurred from year
to year (Ellis et al. 1994). Of greater concern, however, is
the variation in expression levels and patterns that occur
between individual transformants containing the same
construct. In addition to the variation in the overall levels
and patterns of transgene expression, current molecular
understanding of transgene regulation in plants is at a relatively elementary level. Genetic engineering for sterility
requires very precise control over a transgene to interrupt
and terminate flowering. There currently is no way to ensure that this transgene will function in all the plants over
the 5 to 10 years a poplar plantation requires to mature.
Additional research is needed to devise such capabilities.
Evolution of Resistant Pest
Biotypes and Emergence of
Secondary Pests
General Considerations
the evolution of insect and microbial biotypes in response to genetically engineered plants has been an area
of concern since the early development of plant transformation technologies (Gould 1988). The same 3 questions
posed for gene escape, with emphasis on insects, will be
addressed here: 1) Is there assurance that resistant pests
will not evolve? 2) Is there assurance that resistant pests
will not cause environmental harm if they do arise? and 3)
Are there ways for reducing the likelihood and impact of
resistant biotypes?
The potential of pest-biotype evolution is well established by a strong mechanistic foundation and historical
precedent (Brattsten et al. 1986; Forgash 1984; Georghiou
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
and Lagunes-Tejeda 1991; Georghiou and Saito 1983;
McGaughey 1985; Roush and Tabashnik 1990; Tabashnik
1994). Extensive insecticide application, deployment of
resistant cultivars, and even cultural practices have repeatedly selected for resistant insect biotypes (Via 1990). Moreover, the underlying mechanisms have been well
characterized at the molecular, biochemical, physiological, and population levels (Cohan and Graf 1985; EggersSchumacher 1983; Flexon and Rodell1982; Kulkarni and
Hodgson 1984; Muggleton 1982; Mullin and Scott 1992;
Oppenoorth 1984; Roush 1987; Ryan and Byrne 1988;
Skylakakis 1982). Intraspecific differences among gypsy
moths and forest tent caterpillars from different geographic
sources to poplars transgenically expressing Bt have been
observed (figure 1). Variation among these and other forest lepidopterans to exogenously applied Bt is also well
documented (Rossiter et al. 1990; Van Frankenhuyzen et
al. 1995). Adaptations by insects to altered sources of food
plant quality, quantity, and distribution are well documented in natural and managed systems (Singer et al. 1993;
1.0
. . M. DISSTRIA
~ L. DISPAR
_.
(/)
0
a::
0.8
....
z
0
(.)
a::
0.6
0
u.
w
(.)
zw
0.4
a::
w
u.
w
a::
Q..
0.2
0.0
ONT
WI
NS
F51
POPULATION
Figure 1. Intraspecific variation among gypsy moth and
forest tent caterpillar populations to transgenic
poplars expressing a Bt endotoxin gene. Two
forest tent caterpillar, Malacosoma disstria,
populations showed different levels of aversion
from transgenic relative to control foliage. A
similar difference was observed between 2
gypsy moth, Lymantria dispar, populations
(Robinson et al. 1994 )_. ONT=Ontario, Canada;
WI=Wisconsin, USA; NS=Nova Scotia, Canada
and USDA APHIS; F51 =gypsy moths from
laboratory culture at USDA Beneficial Insects
Laboratory, Newark, DE, USA; ns=nonsignificant.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Singer and Parmesan 1993; Via 1990). Thus, herit~ble variation required for gene frequency alteration in response to
selection is sufficient.
It does not matter whether an insect toxin is deployed
through spray application, traditional breeding, or genetic
transformation. For example, the introduction of new plant
varieties is sometimes followed by a population increase
of previously innocuous herbivores to pest levels. Attention should not be limited to plants specifically engineered
for pest resistance, as development of presumably unrelated plant qualities can also alter selective pressures (Raffa
1989). For example, development of new rice varieties for
various agronomic properties led to the emergence of new
pest complexes (Oka and Bahagiaivani 1984). In addition,
engineering plants for altered, increased, or novel secondary products could have multiple consequences. Even
chemical groups that are typically considered "defensive"
can directly benefit adapted herbivores (Bernays and
Woodhead 1982), or indirectly benefit them by reducing
the efficacy of beneficial predators (Codella and Raffa
1995), parasitoids (Campbell and Duffey 1979), and
entomopathogens (Andrews et al. 1980).
Selective pressures imposed by transgenic trees could
be higher than those resulting from insecticide treatments.
Insecticide-resistant biotypes have been rare among forest pests, despite a rather extensive history of synthetic
chemical and Bt application. Tree-feeding herbivores are
not physiologically unique in this regard. Rather, tree-feeding insects are exposed to more variable exposure patterns,
less complete spray deposition, and less frequent treatment, than are agronomic insects. These sources of variability could be lost using transgenics. Depending on
migration rates, neighboring untreated forests could provide refugia for susceptible genotypes, when they are near
to transgenic plantings. In this regard, transgenic poplars
could pose less of a threat of biotype evolution than many
transgenic agronomic crops. Where poplars are planted in
isolation from native refugia, however, such as in irrigated
deserts, this benefit would be minimal.
Usually, the impact of evolved resistance against pesticides, resistant cultivars, and genetically engineered trees,
is limited to loss of efficacy (Raffa 1989). However, 2 general categories of adverse effects could extend beyond
treated plantations: 1) induced insect emigration into
neighboring non-transgenic stands; and 2) lost efficacy of
previously useful tools in non-transgenic stands.
The possibility of between-stand movement raises the
ethical concern that a grower who chooses not to plant
genetically engineered trees may be subjected to immigrants from engineered stands (the same could be true of
most traditionally bred forms of resistance and chemical
pesticides). This risk may be greater in tree than agronomic
crops, because airborne larval dispersion is on average
more important in forest insect life histories. A critical need
in assessing this impact is understanding the relationship
253
Section V Biotechnological Applications
between toxicity and repellency, and in particular, whether
repellency occurs pre- or post- ingestion (Hoy and Head
1995; Ramachandran et al. 1993a, 1993b). This could affect, for example, whether emigrating individuals possess
a slightly higher level of physiological tolerance than the
general population. The available evidence suggests. that
these parameters vary markedly with the insecticidal product, its interaction with plant allelochemicals, and insect
species. We currently lack information on how best to balance the value of behavioral aversion as a resistance-delaying tactic versus its impact on non-treated stands (Gould
1988; Hoy and Head 1995; Johnson and Gould 1992).
The possibility of transgenic plants reducing the utility
of an externally sprayed biopesticide, such as Bt endotoxin, to growers who use Integrated Pest Management
(IPM) approaches based on economic injury levels, raises
a similar ethical concern. This is analogous to neighboring growers applying insecticides on a calendar rather than
density-activated basis. Likewise, integrated plant protection programs could be compromised when biotypes
evolve against transgenic resistances that are based on elevated or altered allelochemicals (again, the same could
be true of some traditional breeding).
Loss of efficacy can be compounded when the mechanism of evolved resistance confers cross-resistance to other
insecticides or resistant-cultivar allelochemicals (Brattsten
et al. 1986; Brattsten 1991). Cross-resistance is a widely
occurring phenomenon that can arise by a number of wellcharacterized physiological mechanisms. For example, the
introduction of the synthetic pyrethroids, derived analogues of Chrysanthemum spp. extracts, encountered rapid
biotype evoiution in regions where the synthetic organochlorine DDT had been widely used. Similarities in the
pharmacological properties of these 2 groups provide a
physiological explanation for cross-resistance, but a priori
considerations based on the unrelatedness of their molecular structures failed to predict these consequences.
Mitigation Strategies
Although resistant biotypes are likely to evolve if
transgenic poplars are deployed without preconceived
resistance management programs, a variety of ameliorating strategies can be used. There are many examples of
effective pest control tactics providing satisfactory control
over many decades. Likewise, naturally evolved plant
defenses provide many examples of long-term stability.
Even among trees, in which host-generation times exceed
those of insects and pathogenic microbes by orders of
magnitude, most members of the host population are protected most of the time (e.g., Edmunds and Alstad 1978;
Whitham 1983). A major principle to emerge independently
from toxicology, plant breeding, and ecology is that the
pattern and intensity of selection, more than the actual
mode of toxicity, most strongly affect biotype evolution
254
rates (Brattsten et al. 1986; Tabashnik and Croft 1982). The
rate, impact, and extent can be greatly reduced by considering features of the target system and by incorporating
heterogeneity at multiple levels of scale. Preconceived resistance management programs now accompany the introduction of many pesticides, as agrichemical
corporations recognize the economic value of protecting
their investments. Likewise, deployment of transgenic
cotton and corn is now accompanied by guidelines prescribing inclusion of non-engineered seed.
Some features of the tree-insect system that can accelerate or retard biotype evolution include the availability of
refugia for susceptible insect genotypes, attributes of the
major pests' physiology, behavior, and ecology, and compatibility of the novel trait with other management tactics. Thus, intensively cultured, short-rotation Populus
plantations pose less risk than forests of long-lived species such as Pseudotsuga. In the latter case, the enormous
differences between pest and host generation times would
greatly reduce the efficacy of any biotype-delaying tactics.
Likewise, the defoliator guild that most strongly impacts
Populus poses less risk than, for example, the scolytids associated with Pinus and Picea. In the latter case, beetle preference for stressed trees limits them to such hosts during
lengthy nonoutbreak periods. Conferring a novel resistance
that was expressed regardless of tree vigor would greatly
alter the selective pressures on bark beetles and possibly
result in more pestiferous behavior (Raffa 1989). No such
relationship appears to regulate population dynamics of
most folivores.
Diversifying tactics, such as mixed block plantings and
crop rotation, are well suited for Populus. A wide range of
hybrid poplar clones are available for deployment, including some that provide rapid growth and resistance against
some key pests (Robison and Raffa 1990, 1994, 1997a,
1997b). Mixed block plantings can also incorporate host
plant tolerance. For example, some of the less resistant
clones against Malacosoma disstria feeding can withstand
considerable defoliation without experiencing severe
growth losses (figure 2).
Protection against multiple pest complexes can be
achieved by integrating traditional and transgenic resistances. The need and potential for this approach are illustrated by the lepidopteran and coleopteran feeding guilds.
Strong resistances against both groups have been identified, but no clones are highly resistant to both (Robison
and Raffa 1994). Continued hybridization or characterization are unlikely to improve this relationship because the
same allelochemicals, specific phenolic glycosides, which
inhibit lepidopterans benefit coleopterans (Bingaman and
Hart 1993; Lindroth and Bloomer 1991; Ramachandran et
al. 1994; Smiley et al. 1985) (figure 3). Understanding these
relationships can help guide molecular strategies. For example, inserting only a coleopteran-active Bt (crylllA,
cryiB) into 'NM6' (P. nigra x P. maximowiczii), and only the
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
~
i?ii
t5
~~
~~
c:li:
~~
w
a:
b
A
DTAC2
12-
NM6
...J
10
<
>
>
a:
:::)
en
NC5331
8
60
B
~
40
~
~
~
~
·o---- -------0
B
oe
NC11432
20
NC11445
6
NC11382
~--
~c;;
80
14- NC11505
NC5262
0
4
50
NC11396
60
70
80
90
NC11004
2
PHENOLICS (ug/mg)
NC5377
~
Chrysomela scripta
--o--
Ma/acosoma disstria
NC527t
0
22
I
0
2
4
6
8
10
12
14
55
B
a
TOLERANCE TO DEFOLIATION
:E.
(RANK ORDER)
::!!
20
w
i=S
!z·@w
18
&ti
16
B
--- ---
______ ..o
50
~
45
:::E
40
!z;;
w .~
35
g::e
30
Gj
c
w
t)
~VI
i= .!
~"tl
Figure 2. Variation in host resistance against defoliation by
forest tent caterpillar (FTC) larvae and tolerance
to a controlled level of artificial defoliation among
hybrid poplar clones. Some clones are relatively
unable to prevent defoliation but are highly
tolerant if it occurs (from Robison & Raffa
1997a). cl. 'DTAC2' (Populus deltoides var.
angulata x P. x berolinensis); cl. 'NC5262' (cl.
'NE387') (P. balsamifera var. subcordatal
candicans x P. x berolinensis); cl. 'NC5271 (cl.
'NE19') (P. nigra var. charkowiensis x P. nigra
var. caudina); cl. 'NC5331' (cl. 'NE299') (P. nigra
var. betulifolia x P. trichocarpa); cl. 'NC5377' (cl.
'Wisconsin #5') (P. deltoides x P. nigra); cl.
'NC11004' (P. deltoides cv. 'Siouxland'); cl.
'NC11382' (cl. 'NE27') (P. nigra var.
charkowiensis x P. x berolinensis); cl. 'NC11396'
( cl. 'NE49') (P. maximowiczii x P. x bero/inensis);
cl. 'NC11432' (cl. 'NE252') (P. deltoides var.
angulata x P. trichocarpa); cl. 'NC11445' (cl.
'NE280'; cl. 'NE157') (P. nigra x P. /aurifolia); cl.
'NC11505' (cl. 'NE388'; cl. 'NE88') (P.
maximowiczii x P. trichocarpa); cl. 'NM6' (cl.
'Max-5') (P. nigra x P. maximowiczil).
Figure 3. Opposing effects of foliar phenolics on 2 defoliating pests attacking hybrid poplars. The cottonwood leaf beetle, Chrysomela scripta, exhibits
high survival and rapid development on clones,
such as 'NM6' (Populus nigra x P. maximowiczit),
which have high phenolic contents. Survival and
development rates are poor on clones, such as
'NC5271' (P. nigra var. charkowiensis x P. nigra
var. caudina), which have low foliar phenolic
concentrations. Conversely, the forest tent
caterpillar, M. disstria, experiences high survival
and rapid growth on 'NC5271 ,' and poor survival
and growth on 'NM6.' (Ramachandran et al.
1994).
lepidopteran-active Bt (cryiA(a)) into 'NC5271' (P. nigra var.
charkowiensis x P. nigra var. caudina), can cut in half the
number of genetically engineered trees needed to express
any 1 trait, yet still provide full protection against both
pests (table 1). Different forms of resistances can also be
combined with transgenic traits. For example, the clones
'NE332' (P. simonii x P. x berolinensis) and 'NC11382' (P. nigra var. charkowiensis x P. x berolinensis) show resistances to
M. disstria, but these defenses are based on foliar phenolic
glycosides and bud resins, respectively (table 2)
(Ramachandran etal.1994; Robison and Raffa 1997a). Such
combinations can increase heterogeneity because Bt inter-
acts with different phytochemical groups differently
(Appel and Schultz 1994; Hwang et al. 1995).
Heterogeneity can be further enhanced by linking expression to wound-inducible promoters. Wound-inducible
expression of inserted genes could simulate the" economic
injury levels" that trigger pesticide applications in Integrated Pest Management systems. That is, a certain level
of insect feeding would be tolerated before expression was
elicited. Opportunities for increasing heterogeneity at this
level are supported by existing variation in inducibility
among poplar clones (table 3). However, preliminary evidence suggests that the sensitivity of existing wound-in
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
n.
m
Gj
Q
b
14
12
50
.
w
25
60
NC5271
70
80
NE332
90
NMS
PHENOLICS (uglmg)
255
Section V Biotechnological Applications
Table 1. Proposed integrated use of traditionally bred and natural plant resistance with genetic engineering to confer
protection against multiple pest complexes. An example using hybrid poplar.
Level of protection against pest group
Lepidoptera
Coleoptera
Source of resistance
Example
Heritable plant defense
{Current)
Clone
'NM6' 1
'NC5271' 2
Bt Endotoxin
crylilA
cryiA(a)
Clone+ Bt
'NC5271 ' 1 + cryiA(a)
'NM6' 2 + cryii/A
Transgenic trait
(Current)
Integrated combination
(Proposed)
1
2
High
Low
Low
High
Low
High
High
Low
High
High
High
High
'NM6' =Populus nigra x P. maximowiczii
'NC5271' = P. nigra var. charkowiensis x P. nigra var. caudina
Table 2. Surces of resistance against forest tent caterpillar, Malacosoma disstria, in 2 hybrid poplar clones
(Robison & Raffa 1997a). Foliage of 'NE332' is less
suitable for forest tent caterpillar larvae than is foliage of
'NC11382.' Bud resins in 'NC11382' are more effective at
immobilizing larvae and preventing access to foliage.
Ratio of M. disstria
performance in tissue
'NC11382' 1fNE332' 2
Insect parameter
Foliage
Buds
Growth rate (mg/day)
Development time
Feeding (2nd instar)
Survival
Weight (mg)
Larval mobility
Weight (mg)
6.5
1.5
19.0
2.3
1.7
0.4
0.7
=
1
'NC11382' Populus nigra var. charkowiensis x P. x
berolinensis
2
'NE332' P. simonii x P. x berolinensis
=
Table 3. Clonal variation in Populus inducibility in response
to forest test caterpillar, Ma/acosoma disstria, feeding
Clone
'NC11382'
'NE332'
Percent forest test caterpillar survival
Constitutive tissue Damaged tissue
90
85
85
49
Source: Robison & Raffa (1997a)
can sometimes provide protection nearly equivalent to
treating entire plants (table 4). This approach is most suitable when insects that prefer productive tissues can also
tolerate other foliage. Within-plant spatial variation in
allelochemistry occurs commonly among naturally coevolved plant defenses. For example, in Populus, phenolics are concentrated in the youngest leaves, with the result
that lepidopteran defoliators feed mostly on older tissue.
The underlying physiological basis for uneven phytochemical distribution is complex, but among the benefits incurred by the host are protection of the most
photosynthetically active tissue, retained ability to translocate carbon to the growing tip, and reduced likelihood
of complete defoliation (Coleman 1986; Meyer and Montgomery 1987). Likewise, pines allocate diterpene resin acids to new not old foliage. Thus, pine sawflies feed on the
older foliage, a habit that only removes photosynthetically
less valuable tissue. At first glance, this might suggest a
high potential for these herbivores to overcome such defenses, but most herbivore species have not evolved this
ability. The evolutionary "choice" in this case is not between overcoming a biochemical barrier and starvation.
Rather, those larvae that did feed on new foliage would
grow less and be less fecund than those on old foliage,
and hence be less competitive.
Table 4. Performance of cottonweed leaf beetle, Chrysomela
scripta, on trees completely or partially treated with Bt.
Treatment
ducible promoters may need to be increased before this
strategy can provide field-level efficacy (Ellis et al. 1996}.
Further heterogeneity could arise from tissue- and temporally- specific expression. Protecting only favored leaves
256
Control
All foliage
Young only
Mature only
No. egg masses
produced
10.25
5.75
3.50
9.75
Source: Ramachandran et al. (1994)
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
Alteration of Multi-Trophic
Processes
Experiences with pesticides, introduced organisms, and
other xenobiotic inputs suggest several means by which
products of genetically engineered plants could affect ecosystem-level processes. Adverse impacts include reduced
populations of predators, parasites, scavengers, pollinators, and endangered or aesthetically valued species. Both
direct and indirect mechanisms have been delineated for
impacts on each of the above, and strong historical data
provides examples for each (e.g., see Caltigirone and
Huffaker 1980; Ehler 1990; Findlay and Jones 1990; Holt
1977; Pimentel1980, 1991; Pimentel and Warneke 1989).
Direct effects of xenobiotics on natural enemies can occur by acute toxicity or biomagnification. In general, genetically engineered trees should be less directly damaging
to natural enemies than are traditional pesticides. One of
the major advantages of transgenic plants is that toxins
can be delivered directly to the herbivore, without broadcast application. The likelihood of biomagnification depends on the gene product. To date, most traits engineered
into plants involve gene products that are rapidly broken
down within the target insects. For example, we are unaware of any instances where predators were directly affected from ingesting prey killed by Bt. Other more stable
gene products, however, could be problematic.
Evaluating potential effects on parasites is more difficult. A major concern is that parasitoids will oviposit in
insects in which they cannot complete development before host death. This could drastically reduce parasite
populations, and thereby release secondary pests. For example, negative effects of plant allelochemicals on parasitoid success are well documented (Barbosa and Saunders
1984; Campbell and Duffey 1979). However, xenobiotics
can sometimes benefit parasitoids. For example, Bt application can enhance performance and population densities of the gypsy moth larval parasitoid, Cotesia melanoscela
(Weseloh et al. 1983). In this example, the delayed growth
rates caused by Bt apparently increase the period during
which surviving early instars are vulnerable to parasitism. However, interpreting such results is complicated.
For example, Johnson and Gould (1992) have argued that
synergism between genetically engineered resistance and
parasitoids could accelerate biotype evolution and its resultant hazards. That is, if exposure to a particular product increases the likelihood of ultimate mortality, selection will
more strongly favor tolerance against the predisposing agent.
But in some cases, increased parasitoid densities resulting
from synergism could subsequently exert mortality independent of the xenobiotic. More research is required to
better quantify and partition these multiple effects.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Som.e indirect effects following the introduction of any
new organism are inevitable. In some cases, an introduced
biological control agent may competitively displace other
parasitoids, yet provide less overall control (Ehler and Hall
1982). This most commonly occurs where a non-host factor (e.g., nectar) is limiting. In the case of transgenic poplars, we see no readily apparent mechanism by which this
could occur. Likewise, major effects on aesthetically valued
species outside the genetically engineered plantations seem
unlikely. Threats to endangered invertebrates arise primarily from habitat destruction, so environmental concerns
would be better addressed if they included potential consequences of extensive Populus cultivation per se, rather than
just the transgenic approaches taken to protect them.
Perhaps the least understood component of risk assessment concerns potential adverse affects on soil organisms.
Risk assessment in this area is especially difficult because
our basic understanding of nutrient cycling, detritivore taxonomy and ecology, and soil biochemical and biophysical
processes is limited. Historical precedents of canopy inputs affecting soil processes include anthropogenic inputs,
such as pesticides and other pollutants, and natural products such as foliar nutrients and lignins, allelochemicals
from induced foliage, and insect frass (Aber et al. 1990;
Cates et al. 1995; Horner et al. 1988; Mattson and Addy
1975; Melillo et al. 1982; Sugai and Schimel1993).
Whether the effects of transgenic leaf litter are nonexistent, short- or long- term, and have point or non-point effects, depends on the gene product and soil properties.
Products most commonly suggested for transformation into
Populus have relatively high specificity, which reduces risk.
For example, various Bt endotoxins are specific to Lepidoptera, Diptera, or Coleoptera. The first 2 groups do not
appear to pose significant concerns; Lepidoptera exert a relatively minor role in litter decomposition, and genes conferring protection against Diptera are unnecessary in Populus.
Beetles, however, are important components of the soil fauna,
functioning as decomposers, vectors of beneficial microbes,
and predators on a variety of potentially injurious arthropods
and fungi. Thus, introduction of Coleoptera-active Bt or proteinase inhibitor poses some concern. Gene products with
relatively general activity could be more problematic. For
example, proteinase inhibitors can sometimes affect relatively
diverse taxa and require more detailed evaluation.
Novel gene products could be altered by soil biochemical
and biophysical processes, as occurs with synthetic materials (Angle 1994). Consideration of genetically engineered
organisms must extend beyond the actions of the gene products themselves and include studies of breakdown products.
The consequences of a stable gene product must also be considered. Although a protein may not be toxic to soil organisms at the levels present in a single leaf, buildup in the soil
over a season or many years may pose a problem. Such residual effects are difficult to predict because stability is affected by factors such as soil pH, nutrient content, rainfall,
257
Section V Biotechnological Applications
and temperature. The stability of proteins may also be altered within the plant and vary during the year.
Tactics for reducing potential risks to natural enemies and
detritivores relate to the inherent properties of the gene products themselves and their expression. In general, environmental risk can be minimized when these products and their
derivatives are specific to the target insect, of short duration, and exposed in a spatially and temporally limited pattern. Existing approaches to toxicological evaluation are
available for such analyses. However, further theoretical
development is required before optimal relationships between transgenic plants and parasitoids can be devised.
Conclusion
Some serious environmental concerns must be weighed
against the potential benefits of genetically engineered
Populus. Risk assessment can be improved by focusing on
the most likely sources of environmental harm as opposed
to generic listings of all hypothetical outcomes. As stated
by previous authors from a broad range of backgrounds,
emphasis should be placed on how gene products will
interact with ecosystems not how these products arose.
Conversely, the notion that genetic engineering has somehow been singled out for unique environmental scrutiny
should be dispelled because there is a long history of
guidelines and regulations lim.iting other insect control.
tactics including synthetic and naturally derived insecticides, biological control agents, insect growth regulators,
antifeedants, and even cultural control (e.g., Charudattan
and Browning 1992; Coulson and Soper 1989; Howarth
1991; Mcevoy 1996; Miller 1990; Samways 1988; Upholt
1985; USDA FS 1995).
A specific risk merits concern where its potential is supported by established mechanisms and relevant precedent.
These criteria are met for several potential threats arising
from 3 general categories of risk: 1) escape of engineered
germplasm; 2) evolution of resistant biotypes; and 3) alteration of multi-trophic processes. Underlying mechanisms for
each of these have been well established from multiple disciplines and across molecular through community levels.
Risks can be prioritized as to whether: 1) they would be localized or affect adjacent ecosystems; 2) environmental harm
would depend on continued deployment or be self-perpetu- ·
ating; and 3) potential ameliorating tactics are available.
Table 5 summarizes the major anticipated risks, general
mechanisms by which they might occur, historical precedents
from which valuable lessons can be applied, and possible
preventative strategies. Four points emerge from this overview. First, there is a need for proactive research on the likelihood of various environmental hazards and tactics for
offsetting them. Second, interdisciplinary approaches are
essential. Many of the challenges associated with plant genetic engineering may be identified from ecological perspectives, yet have fundamentally molecular solutions and vice
versa (Raffa 1989).It is especially important that integrative
collaborations function throughout the entire discovery and
development process, rather than in sequential fashion. Sequential approaches fail to fully synergize the expertise that
enhances efficacy and environmental safety and are likely
to generate rivalries from differing vested interests. Third,
none of the risks appears unmanageable if appropriate molecular, physiological, ecological, and management strategies are employed in a cohesive fashion. Fourth, Populus
provides a particularly suitable model for research and deployment. There is a strong knowledge base from genetic,
physiological, ecological, and production perspectives, and
a need for traditional and emerging forest products that can
be economically produced by this genus.
Table 5. Summary of environmental risks, mechanisms, precedents, and preventative strategies for genetically. engineered
·
insect resistant Populus.
Risk
Mechanisms
Precedents
Prevention
Escape
Pollen transfer; hybridization
vegetative materials
Introduced
pests
Sterility; site management
early harvest
Resistant biotypes
Altered selection pressures;
release from competitors,
dispersion, cross resistance
Pesticides;
resistant cultivars
Variable & opposing selective
pressures; between- & within-plant
mosaics, tissue-, temporal- & woundspecific expression
Altered tritrophic
processes
Direct & indirect affects on
beneficial species; effects of
gene products on nutrient cycling
Introduced pests;
pesticides;
pollutants
Specificity of gene products and
breakdown products; rapid
environmental turnover of gene
products; monitoring
258
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
Acknowledgments
''W
This work was supported by the University of Wisconsin-Madison College of Agricultural and Life Sciences,
Hatch, the UW-Madison Graduate School, McintireStennis, and the Consortium for Plant Biotechnology.
Literature Cited
Abbott, R.J. 1994. Ecological risks of transgenic crops.
Trends in Ecology and Evolution. 9: 280-282.
Aber, J.D.; Melillo, J.M.; McClaugherty, C.A.1990. Predicting long-term patterns of mass loss, nitrogen dynamics, and soil organic matter formation from initial fine
litter chemistry in temperate forest ecosystems. Can. J.
Bot. 68: 2201-2208.
Andrews, R.E.; Parks, L.W.; Spence, K.D. 1980. Some effects of Douglas fir terpenes on certain microorganisms.
Appl. Environ. Microbial. 40: 301-304.
Angle, J .S. 1994. Release of transgenic plants: Biodiversity
~nd population -level considerations. Mol. Ecol. 3:45-50.
Appel, H.M.; Schultz, J.C. 1994. Oak tannins reduce effectiveness of thuricide (Bacillus thuringiensis) in the gypsy
moth (Lepidoptera: Lymantriidae). J. Econ. Entomol. 87:
1736-1742.
Barbosa, P.; Saunders, J.A. 1984. Plant allelochemicals:
Linkages between herbivores and their natural enemies.
Recent Advances in Phytochemistry. 19: 107-136.
Bergelson, J. 1994. Changes in fecundity do not predict
invasiveness- a model study of transgenic plants. Ecology. 75: 249-252.
Bemays, E.A.; Woodhead, S. 1982. Plant phenols utilized
as nutrients by a phytophagous insect. Science. 216: 201203.
.
Bingaman, B.R.; Hart, E.R. 1993. Clonal and leaf age variation in Populus phenolic glycosides: Implications for
host selection by Chrysomela scripta (Coleoptera:
Chrysomelidae). Environ. Entomol. 22: 397-403.
Brattsten, L.B. 1991. Bioengineering of crop plants andresistant biotype evolution in insects. Counteracting coevolution. Archives Insect Biochem. & Physiol. 17:
253-267.
Brattsten, L.B.; Holyoke, C.W., Jr.; Leeper, J.R.; Raffa, K.F.
1986. Insecticide resistance: Challenge to pest management and basic research. Science. 231: 1255-1260.
Cates, R.G.; Terry, R.; Zou, J.; Wood, S.G. 1995. The effect
of tannins and phenolics from Alaskan taiga species on
nitrogen fixation in Alnus tenuifolia. Bull. Torrey Bot.
Club. Submitted.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Caltagirone, L.E.; Huffaker, C.B. 1980. Benefits and risks
of using predators and parasites for controlling pests.
In: Lundholm, B.; Stackerus, M., eds. Environmental protection and biological forms of control of pest organisms. Ecol. Bull. (Stockholm) 32: 103-109.
Campbell, B.C.; Duffey, S.S. 1979. Tomatine and parasitic
wasps: potential incompatibility of plant antibiosis with
biological control. Science. 205: 700-702.
Charudattan, R.; Browning, H.W., eds. 1992. Regulations
and guidelines: Critical issues in biological control, Proceedings of a USDA/CSRS national workshop. Institute of Food and Agric. Sciences, Univ. Florida,
Gainesville, FL, U.S.A.
Codella, S.G.; Raffa, K.F. 1995. Host plant influence on
chemical defense in conifer sawflies (Hymenoptera:
Diprionidae). Oecologia. 104: 1-11.
Cohan, F.M.; Graf, J. 1985. Latitudinal cline in Drosophila
melanogaster for knockdown resistance to ethanol fumes
and for rates of response to selection for further resistance. Evolution. 39: 278-293.
Coleman, J. 1986. Leaf development and leaf stress: Increased susceptibility associated with sink-source transition. Tree Physiol. 2: 289-299.
Coulson, J.R.; Soper, R.S. 1989. Protocols for the introduction of biological control agents in the US. In: Kahr, R.P.,
ed. Plant protection and quarantine. Boca Raton, FL,
U.S.A.: CRC Press: 1-35.
Dale, P.J. 1994. The impact of hybrids between genetically
modified crop plants and their related species: General
considerations. Mol. Ecol. 3: 31-36.
Dale, P.L.; Irwin, J.A .. 1995. The release of transgenic plants
from containment, and the move towards their widespread use in agriculture. Euphytica. 85: 425-431.
de Zoeten, G.A. 1991. Risk assessment: Do we let history
repeat itself? Phytopathology. 81: 585-586.
Edmunds, G.R., Jr.; Alstad, D.N. 1978. Coevolution in insect herbivores and conifers. Science. 199: 941-945.
Eggers-Schumacher, H.A. 1983. A comparison of the reproductive performance of insecticide-resistant and susceptible clones of Myzus persicae. Entomol. Ex per. Applic.
34: 301-307.
Ehler, L.E. 1990. Environmental impact of introduced biological-control agents: Implications for agricultural biotechnology. Risk assessment in agricultural
biotechnology: Proc. Internat. Conference. 85-96.
Ehler, L.E.; Hall, R.W. 1982. Evidence for competitive exclusion of introduced natural enemies in biological control. Environ. Entomol. 11: 1-4.
Ellis, D.O.; Rintamaki-Strait, J.; Francis, K.; Kleiner, K.;
Raffa, K.; McCown, B. 1996. Transgene expression in
spruce and poplar: From the lab to the field. In:
Ahuja, M.R.; Boerjan, W.; Neale, D.B., eds. Somatic
cell genetics and molecular genetics of trees.
Dordrecht, The Netherlands: Kluwer Academic Publishers: 159-163.
259
Section V Biotechnological Applications
Ellis, D.O.; Rintamaki-Strait, J.; Wraith, S.; McCown, B.;
Kleiner, K.; Raffa, K.F. 1994. Expression of Introduced
Genes in Trees: Practical and Environmental Considerations. In: Proceedings of the TAPPI 1994 biological sciences symposium. 1994 Oct. 3-6; Minneapolis, MN,
U.S.A.: 139-146.
Ellis, W.R.; Ham, G.E.; Schmidt, E.L. 1984. Persistence and
recovery ofRhizobium japonicum inoculum in a field soil.
Agron. J. 76:573-576.
Falk, B.W.; Bruening, G. 1994. Will transgenic crops generate new viruses and new diseases? Science. 263:
1395-1396.
Federal Insecticide, Fungicide, & Rodenticide Act. 1978
(And subsequent amendments). Environmental Protection Agency.
Findlay, S.; Jones, C.G. 1990. Exposure of cottonwood
plants to ozone alters subsequent leaf decomposition.
Oecologia. 82: 248-250.
Flexon, P.B.; Rodell, C.F. 1982. Genetic recombination and
directional selection for DDT resistance in Drosophila
melanogaster. Nature. 298:672-674.
Forgash, A.J. 1984. History, evolution, and consequences
of insecticide resistance. Pest. Biochem. and Physiol. 22:
178-186.
Frederick, R.J.; Egan, M. 1994. Environmentally compatible applications of biotechnology. BioScience. 44:
529-535.
Fuester, R. W. 1993. Quarantine issues affecting biological
control. In: McDonald, R.C.; Harper, J.D.; Dickerson,
W.A., eds. Biological control: Developing strategies for
the 90's. Proceedings of a national conference. 1991 April
8-11. Raleigh, NC, U.S.A.: Plant Industry Division, Plant
Protection Section, North Carolina Department of Agriculture: 22-26.
Georghiou, G.P.; Lagunes-Tejeda, A., eds. 1991. The occurrence of resistance to pesticides in arthropods: an index
of cases reported through 1989. Rome: Food and Agriculture Organization of the United Nations.
Georghiou, G.P.; Saito, T. 1983. Pest resistance to pesticides.
New York, NY, U.S.A.: Plenum Press: 809 p.
Giampietro, M. 1994. Sustainability and technological development agriculture - a critical appraisal of genetic
engineering. BioScience. 44: 677-689.
Gould, F. 1988. Evolutionary biology and genetically engineered crops. BioScience. 38: 26-33.
Harris, P. 1985. Biocontrol and the law. Bull. Entomol. Soc.
Can. 17: 1:
Hinkle, M.K. 1993. Environmental issues of biological
control regulation. In: McDonald, R.C.; Harper, J.D.;
Dickerson, W.A., eds. Biological control: Developing
strategies for the 90's. Proceedings of a national conference. 1991 April 8-11. Raleigh, NC. Raleigh, NC,
U.S.A.: Plant Industry Division, Plant Protection
Section, North Carolina Department of Agriculture:
13-21.
260
Holt, R. 1977. Predation, apparent competition, and the
structure of prey communities. Th.eor. Popul. Bioi. 12:
197-229.
Homer, J.D.; Gosz, J.R.; Cates, R.G. 1988. The role of carbon-based plant secondary metabolites in decomposition in terrestrial ecosystems. Amer. Naturalist. 132:
869-883.
Howarth, F.G. 1991. Environmental impacts of classical
biological control. Ann. Rev. Entomol. 36: 485-509.
Hoy, C.W.; Head, G. 1995. Correlation between behavioral
and physiological responses transgenic potatoes contafning Bacillus thuringiensis delta-endotoxin in
Leptinotarsa decemlineata (Coleoptera: Chrysomelidae).
J. Econ. Entomol. 88: 480-486.
Hubbes, M. 1993. Impact of molecular biology on forest
pathology- a literature review. Europ. J. Forest Pathol.
23: 201-217.
Hutton, P. 1992. Regulation of microbial biological pest
control agents by the Environmental Protection Agency.
In: Charudattan, R.; Browning, H.W., eds. Regulations
and guidelines: Critical issues in biological control, Proceedings of a USDA/CSRS national workshop. Institute of Food and Agric. Sciences. Gainesville, FL, U.S.A.:
Univ. Florida: 25-30.
Hwang, S.Y.; Lindroth, R.L.; Montgomery, M.E.; Shields,
K.S. 1995. Aspen leaf quality affects gypsy moth (Lepidoptera: Lymantriidae) susceptibility to Bacillus
thuringiensis. J. Econ. Entomol. 88: 278-282.
Jepson, P.C.; Croft, B.A.; Pratt, C.E. 1994. Test systems to
determine the ecological risks posed by toxins release
from Bacillus thuringiensis genes in crop plants. Mol.
Ecol. 3: 81-89.
Johnson, M.T.; Gould, F. 1992. Interaction of genetically
engineered host plant resistance and natural enemies
of Heliothis virescens (Lepidoptera, Noctuidae) in Tobacco. Environ. Entomol. 21: 586-597.
Kareiva, P.; Morris, W.; Jacobi, C.M. 1994. Studying and
managing the risk of cross-fertilization between
transgenic crops and wild relatives. Mol. Ecol. 3: 15-21.
Kleiner, K.W.; Ellis, D.O.; McCown, B.H.; Raffa, K.F. 1995.
Field evaluation of transgenic poplar expressing Bacillus thuringiensis d-endotoxin gene against forest tent caterpillar (Lepidoptera: Lasiocampidae) and gypsy moth
(Lepidoptera: Lymantriidae). Environ. Entomol. 24:
1358-1364.
Kulkarni, A.P.; Hodgson, E. 1984. The metabolism of insecticides: the role of monooxygenase enzymes. Ann.
Rev. Pharmacal. Toxicol. 24: 19-42.
Liebhold, A.M.; MacDonald, W.L.; Bergdahl, D.; Mastro,
· V.C. 1995. Invasion by exotic forest pests: A threat to
forest ecosystems. Forest Sci. Monogr. 30.49 p.
Lindroth, R.L.; Bloomer, M.S. 1991. Biochemical ecology of the forest tent caterpillar: Responses to dietary
protein and phenolic glycosides. Oecologia. 86: 408413.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
Lodge, D.M. 1993. Biological invasions - lessons for ecology. Trends in Ecology & Evolution. 8: 133-137.
Mariani, C.; DeBeuckleer, M.; Truellner, J.; Leemans, ].;
Goldberg, R.B. 1990. Induction of male sterility in plants
by a chimeric ribonuclease gene. Nature. 347: 737-741.
Mattson, W.J.; Addy, N.D. 1975. Phytophagous insects as
regulators of forest primary production. Science. 190:
515-522.
McCown, B.H.; McCabe, D.E.; Russell, D.R.; Robison, D.].;
Barton, K.A.; Raffa, K.F. 1991. Stable transformation of
Populus and incorporation of pest resistance by electrical discharge particle acceleration. Plant Cell Reports.
5:590-594.
Mcevoy, P.B. 1996. Host specificity and biological pest control. How well is research on host specificity addressing the potential risks of biological control? BioScience.
46: 401-405.
McGaughey, W.H. 1985. Insect resistance to the biological
insecticide Bacillus thuringiensis. Science. 229: 193-195.
McGranahan, G.M.; Leslie, C.A.; Uratsu, S.L.; Martin, L.A.;
Dandekar, A.M. 1988. Agrobacterium-mediated transformation of walnut somatic embryos and regeneration of
transgenic plants. Bio/Technol. 6: 800-804.
Melillo, J.M.; Aber, J.D.; Mura'tore, ].F. 1982. Nitrogen and
lignin control of hardwood leaf litter decomposition dynamics. Ecology. 63: 621-626.
Meyer, G.A.; Montgomery, M.E. 1987. Relationships between leaf age and the food quality of cottonwood foliage for the gypsy moth, Lymantria dispar. Ecology. 72:
527-532.
Miller, J.C.D. 1990. Field assessment of the effects of amicrobial pest control agent on nontarget Lepidoptera.
Amer. Entomol. 36: 135-139.
Moamed, H.W.; Ellis, W.R.; Schmidt, E.L. 1984. Rhizosphere
responses as a factor in competition among three
serogroups of indigenous Rhizobium japonicum for inoculation of field-grain soybeans. Applied & Environ.
Microbial. 47: 607-612.
Mooney, H.A.; Drake, J.A., eds. 1986. Ecology of biological invasions of North America and Hawaii. New York,
NY, U.S.A.: Springer-Verlag. 321 p.
Morgan, G.M. 1993. Risk analysis and management. Sci.
Amer. 269: 32-41.
Muggleton, J. 1982. A model for the elimination of insecticide resistance using heterozygous disadvantage. Heredity. 49: 247-251.
Mullin, C.A.; Scott, J.G., eds. 1992. Molecular mechanisms
of insecticide resistance: diversity among insects. Washington, D.C., U.S.A.: American Chemical Society: 322 p.
National Academy of Sciences, 1987. Committee on the
Introduction of Genetically Engineered Organisms into
the Environment. Washington, D.C., U.S.A.: National
Academy Press.
National Academy of Sciences, 1989. Field testing genetically-modified organisms. Washington, D.C., U.S.A.:
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
National Academy Press.
Oka, F.N .; Bahagiaivani, H. 1984. Development and management of a new brown plant hopper (Nilaparvata
leguna Stal) biotype in North Sumatra, Indonesia. Indonesia. Contributions from the Central Research Institute for Food Crops Bogor. 71: 1-33.
Oppenoorth, F.J. 1984. Biochemistry of insecticide resistance. Pestic. Biochem. Physiol. 22: 187-193.
Pimentel, D. 1980. Environmental risks asspciated with
biological controls. In: Lundholm, B.; Stackerud, M.,
eds., Environmental protection and biological forms of
control of pest organisms. Ecol. Bull. (Stockholm) 31:
11-24.
Pimentel, D., ed. 1991. Handbook of pest management in
agriculture. Boca Raton, FL, U.S.A.: CRC Press.
Pimentel, D.; Hunter, M.S.; LaGro, ].A.; Efroymson, R.A.;
Landers,J.C.D.; Mervis, F.T.; McCarthy, C.A.; Boyd, A.E.
1989. Benefits and risks of genetic engineering in agriculture. BioScience. 39: 606-614.
Pimentel, D.; Warneke, A. 1989. Ecological effects of manure, sewage, sludge and other organic wastes on arthropod populations. Agr. Zool. Rev. 3: 1-32.
Price, P.W.; Bouton, C.E.; Gross, P.; McPheron, B.A.; Thompson, J.N .; Weis, A.E. 1980. Interactions among three
trophic levels: Influence of plants on interactions between insect herbivores and natural enemies. Annual
Rev. Ecol. & System. 11: 41-65.
Raffa, K.F. 1989. Genetic engineering of trees to enhance
resistance to insects: Evaluating the risks of biotype evolution and secondary pest outbreak. BioScience. 39: 524534.
Raffa, K.F.; McCown, B.; Ellis, D.; Ramachandran, R.;
Robison, D.J. 1993. Prospects and approaches to genetically engineered pest resistance in energy trees. In: Hall,
R.B.; Hanna, R.D.; Nyong' o, R.N., eds. Proc. Internat.
Energy Agency 1991 Joint meeting of task force V groups
on exchange of genetic materials and joint trials of Alnus,
Populus, & Salix. 1991 August 22-27; Ames, lA, U.S.A.:
Ames, lA, U.S.A.: Iowa State University: 98-106.
Ramachandran, R.; Raffa, K.F.; Bradley, D.; Miller, M.; Ellis,
D.; McCown, B. 1993a. Activity of an insecticidal protein from Bacillus thuringiensis subsp. thuringiensis HD290-1 strain to coleopteran and lepidopteran defoliators
of poplars. Environ. Entomol. 22: 190-196.
Ramachandran, R.; Raffa, K.F.; Miller, M.J.; Ellis, D.E.;
McCown, B.H. 1993b. Behavioral and physiological responses of spruce budworm and fall webworm larvae
to Bacillus thuringiensis Cry 1A(a) toxin. Environ.
Entornol. 22: 197-211.
Ramachandran, R.; McCown, B.H.; Ellis, D.E.; Raffa, K.
1994. Biotechnology and host plant resistance: Benefits,
risks, strategies for development and deployment. In:
Ananthakrishnan, T.N .; Raman, A., eds. Chemical ecology of phytophagous insects. New York: International
Science Publishers: 275-288.
261
Section V Biotechnological Applications
Regal, P.J. 1994. Scientific principles for ecologically based
risk assessment of transgenic organisms. Mol. Ecol. 3:
5-13.
Robison, D.J.; McCown, B.H.; Raffa, K.F. 1994. Responses
of gypsy moth (Lepidoptera: Lymantriidae) and forest
tent caterpillar (Lepidoptera: Lasiocampidae) to
transgenic poplar, Populus spp., containing a Bacillus
thuringiensis d-endotoxin gene. Environ. Entomol. 23:
1030-1041.
Robison, D.J.; Raffa, K.F. 1990. Hybrid poplar productivity, and suitability for the forest tent caterpillar: A framework for evaluation. Proc. 1989 aspen symposium; 1989
July 25-27; Duluth, MN, U.S.A. Gen. Tech. Rep. NC-140.
St. Paul, MN, U.S.A.: U.S. Dept. of Agriculture, Forest
Service: 155-162.
Robison, D.J.; Raffa, K.F. 1994. Characterization of hybrid
poplar clones for resistance to the forest tent caterpillar.
Forest Science. 40: 686-714.
Robison, D.J.; Raffa, K.F. 1997a. Effects of constitutive and
inducible traits of hybrid poplars on forest tent caterpillar feeding and population ecology. Forest Science.
In press.
Robison, D.J.; Raffa, K.F. 1997b. Productivity, drought tolerance, and pest status of hybrid Populus: Tree improvement and silvicultural implications. Biomass and
Bioenergy. In press.
Rossiter, M.; Yendol, W.G.; DuBois, N.R. 1990. Resistance
of Bacillus thuringiensis in gypsy moth (Lepidoptera:
Lymantriidae): genetic and environmental causes. J.
Econ. Entomol. 83: 2211-2218.
Roush, R. T. 1987. Ecological genetics of insecticide and
acaricide resistance. Ann. Rev. Entomol. 32:361-380.
Roush, R.T.; Tabashnik, B.E., eds. 1990. Pesticide resistance
in arthropods. New York: Chapman and Hall: 303 p.
Ryan, M.F.; Byrne, 0. 1988. Plant-insect coevolution and
inhibition of acetylcholinesterase. J. Chern. Ecol. 14:
1965-1975.
Samways, M.J. 1988. Classical biological control and insect conservation: Are they compatible? Environ.
Conserv. 15: 349-354.
Seidler, R.J .; Levin, M. 1994. Potential ecological and nontarget effects of transgenic plant gene products on agriculture, silviculture, and natural ecosystems: general
introduction. Mol. Ecol. 3: 1-3.
Simmonds, N.W., ed. 1976. The evolution of crop plants.
London, England: Longmans: 339 p.
Simberloff, D. 1985. Predicting ecological effects of novel
entities: evidence from higher organisms. In: Halverson,
H.O.; Pramer, D.; Rogul, M., eds. Engineered organisms
in the environment: Scientific issues. Washington, D.C.,
U.S.A.: Amer. Soc. Microbial.: 152-161.
Singer, M.C.; Parmesan, C. 1993. Sources of variation in
patterns of plant-insect interactions association. Nature.
361: 251-253.
Singer, M.C.; Thomas, C.D.; Parmesan, C. 1993. Rapid hu-
262
man-induced evolution of insect diet. Nature. 366: 681683.
Skylakakis, G. 1982. Epidemiological factors affecting the
rate of selection of biocide-resistant genotypes of plant
pathogenic fungi. Phytopath. 72: 271-273.
Smiley, T.J.; Hom, J.M.; Rank, N.E. 1985. Ecological effects
of salicin at three trophic levels: New problems from
old adaptations. Science. 229: 649-651.
Strauss, S.H.; Rottmann, W.H.; Brunner, A.M.; Sheppard,
L.A. 1995. Genetic engineering of reproductive sterility
in forest trees. Mole. Breeding. 1: 5-26.
Strauss, S.H.; Howe, G.T.; Goldfarb, B. 1991. Prospects for
genetic engineering of insect resistance in forest trees.
Forest Ecology and Management. 43: 181-209.
Sugai, S.F.; Schimel, J.P. 1993. Decomposition and biomass
incorporation of 14C-labeled glucose and phenolics in
taiga forest floor: effect of substrate quality, successional
state, and season. Soil Bioi. Biochem. 25: 1379-1389.
Tabashnik, B.E. 1994. Evolution of resistance to Bacillus
thuringiensis. Annual Rev. Entomol. 39: 47-79.
Tabashnik, B.E.; Croft, B.A. 1982. Managing pesticide resistance in crop-arthropod complexes: Interactions between biological and operational factors. Environ.
Entomol. 11: 1137-1144.
Tiedje, J.M.; Colwell, R.K.; Crossman, Y.L.; Hodson, R.E.;
Lenski, R.E.; Mack, R.N.; Regal, P.J. 1989. The planned
introduction of genetically engineered organisms: Ecological considerations and recommendations. Ecology.
70: 298-315.
Timmons, A.M.; Charters, Y.M.; Dubbels, S.J.; Wilkinson,
M.J. 1995. Assessing the risks of wind pollination from
fields of genetically modified Brassica napus ssp. oleifera.
Euphytica. 85: 417-423.
Timmons, A.M.; Charters, Y.M.; Crawford, J.W.; Burn, D.;
Scott, S.E.; Dubbels, S.J.; Wilson, N.J.; Robertson, A.;
O'Brien, E.T.; Squire, G.R.; Wilkinson, M.J. 1996. Risks
from transgenic crops. Nature. 380:487.
Tolin, S.A.; Vidaver, A.K. 1989. Guidelines and regulations
for research with genetically modified organisms: a view
from Academe. Annual Rev. Phytopath. 27: 551-581.
Upholt, W.M. 1985. The regulation of pesticides. In:
Mandara, N.B., ed. Handbook of natural pesticides.
Methods, theory, product, and detection. Boca Raton,
FL, U.S.A.: CRC Press: 273-295. Vol. 1.
Urban, D.J.; Cook, N.J. 1986. Hazard Evaluation Division
standard evaluation procedure: Ecological risk assessment. Washington, D.C., U.S.A.: US EPA Office of Pesticides Programs.
US Congress, Office of Technology Assessment. 1993.
Harmful non-indigenous species in the United States.
OTA-F-565. Washington, D.C., U.S.A.: Govt. Printing
Office.
USDA FS. 1991. Pest risk assessment of the importation of
larch from Siberia and the Soviet far east. USDA Forest
Service/Miscellaneous Publication no. 1495: S-1-L-6.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
Environmental Risk Assessment and Deployment Strategies for Genetically Engineered Insect-Resistant Populus
USDA FS. 1995. Record of decision. Final environmental
impact statement for the management of the redcockaded woodpecker and its habitat on national forests in the southern region. Manag. Bull. R8-MB-73.
Van Frankenhuyzen, K.; Nystron, C.W.; Tabashnik, B.E.
1995. Variation in tolerance to Bacillus thuringiensis
among and within populations of the spruce bud worm
(Lepidoptera: Totricidae) in Ontario. J. Econ. Entomol.
88:97-105.
Via, S. 1990. Ecological genetics and host adaptation in
herbivorous insects: The experimental study of evolution in natural and agricultural systems. Annual Rev.
Entomol. 35: 421-446
Weseloh, R.M.; Andreadis, T.G.; Moore, R.E.B.; Anderson,
J.F.; DuBois, N.R.; Lewis, F.B. 1983. Field confirmation
of a mechanism causing synergism between Bacillus
thuringiensis and the gypsy moth parasitoid, Apanteles
melanoscelus. ]. Insect Pathol. 41: 99-103.
USDA Forest Service Gen. Tech. Rep. RM-GTR-297. 1997.
White, W.B.; DeNitto, G.A.; Hanson, J.B.; Bellinger, M.D.;
Russell, J.T.H. 1992. Pest risk assessment on the importation of Pinus radiata and Douglas-fir logs from New
Zealand. Draft Copy: 1-61.
Whitham, T. 1983. Host manipulation of parasites: withinplant variation as a defense against rapidly evolving
pests. In: Denno, R.F.; McClure, M.S., eds. Variable plants
and herbivores in natural and managed systems. New
York: Academic Press: 15-41.
Williamson, M. 1994. Community response to
transgenic plant release: Predictions from British
experience of invasive plants and feral crop plants.
Mol. Ecol. 3: 75-79.
Wilson, R.; Crouch, E.A.C. 1987. Risk assessment and comparison: an introduction. Science. 236: 267-270.
Windle, P.M.; Franz, E.H. 1979. The effects of insect parasitism on plant competition: greenbugs and barley. Ecology. 601: 521-529.
263